5090 mobile GPU is stuck on 95W TGP - is this a driver issue?

[Bill2462]

Hi!

I bought the Asus Strix SCAR 18 G835L laptop with a 5090 mobile for ML research and gaming, and unfortunately, the GPU is suck on the default TGP setting of 95W. So far, the driver seems operational, and PyTorch and other deep learning packages are working correctly. Still, the power issue is annoying because it removes almost half of the performance.

I verified on Windows 11 that the GPU can reach 175W, so it seems to be a driver issue, not a hardware / BIOS problem. Or at least some functionality is broken/missing on Linux but not on Windows.

My system is as follows:

• OS: Manjaro Linux
• Kernel: Linux 6.14.6-2-MANJARO #1 SMP PREEMPT_DYNAMIC Wed, 14 May 2025 14:56:34 +0000 x86_64 GNU/Linux
• Hardware: GPU 0: NVIDIA GeForce RTX 5090 Laptop GPU (UUID: GPU-8ea15d32-2110-b651-5439-4c790e7e5e3a)
• driver: 570.144-12 (from linux614-nvidia-open package)

The problem is that when I try to change the limit, the nvidia-smi says it is unsupported.

So sudo nvidia-smi -pl 175 outputs: Changing the power management limit is not supported for GPU: 00000000:02:00.0

Does anyone have a clue what can be going wrong here?

I attached the bug report file if it can be of any help: nvidia-bug-report.log.gz

I would be grateful for any suggestions. I tried installing a proprietary driver, but it is not working at all. The GPU is not detected by the nvidia-smi tool.

[RENyunfan]

Could someone give some suggestions? similar problem

[arisu3]

Shot in the dark, but could you post the output of nvidia-smi -q?

[Bill2462]

Update:

The problem seems to have been resolved at some point. My current version of the driver is 575.64.05-7 and the limit is now 175W, which is the max TDP for this GPU. I also updated the kernel to 6.16.4-21 but I suspect that updating the driver, not the kernel resolved the issue. Edit: the driver is the open source version, so linux616-nvidia-open for Manjaro and Arch users.

[zletpm]

RTX 5090 Laptop stuck at 95W when eGPU (RTX 5080) is connected — NPCF Dynamic Boost bug

System:

• MSI laptop, Intel Core Ultra 9 275HX
• Internal GPU: RTX 5090 Laptop GPU (24GB) — PCI bus 01:00.0, SUBSYS 14741462
• External GPU: RTX 5080 (Thunderbolt eGPU) — PCI bus 54:00.0, SUBSYS F3221569
• Driver: 32.0.15.9531 (nvpcf.sys 2026-01-13)
• Windows 11 Pro 26200
• MSI Center: Extreme Performance mode, Discrete Graphics Mode

Problem: When the RTX 5080 eGPU is connected, NPCF (nvpcf.sys) assigns Dynamic Boost to the external desktop GPU instead of the internal laptop GPU. This locks the RTX 5090 Laptop at its base TDP of 95W instead of 175W.

Testing performed:

Condition | 5090 enforced power limit -- | -- eGPU disabled in Device Manager | 175W (correct) eGPU enabled | 95W (incorrect) eGPU enabled + NPCF disabled | 95W (no Dynamic Boost at all) eGPU enabled + NPCF cycled (disable/re-enable) | 95W eGPU enabled + MSI Center profile toggled (Extreme/Balanced/ECO via WMI) | 95W nvidia-smi -pl 175 (admin) | "Changing power management limit is not supported" NVML nvmlDeviceSetPowerManagementLimit | "Not Supported"

Conclusion: NPCF is the sole source of Dynamic Boost for the laptop GPU. When two NVIDIA GPUs are present, it targets the wrong one. The desktop eGPU has its own 360W power supply and does not need Dynamic Boost.

Expected behavior: nvpcf.sys should identify the internal laptop GPU by:

• PCI bus topology (01:00.0 = internal, 54:00.0 = Thunderbolt)
• Subsystem vendor ID (1462 = MSI OEM = internal GPU)
• Or ACPI _DSM presence

and assign Dynamic Boost exclusively to the internal GPU.

Workaround: The only working workaround is disabling the eGPU in Device Manager, which defeats the purpose of having one.

This issue also affects RTX 40-series laptops — see egpu.io thread and GeForce Forums thread.

RTX 5090 Laptop stuck at 95W when eGPU (RTX 5080) is connected — NPCF Dynamic Boost bug

System:

MSI laptop, Intel Core Ultra 9 275HX
Internal GPU: RTX 5090 Laptop GPU (24GB) — PCI bus 01:00.0, SUBSYS 14741462
External GPU: RTX 5080 (Thunderbolt eGPU) — PCI bus 54:00.0, SUBSYS F3221569
Driver: 32.0.15.9531 (nvpcf.sys 2026-01-13)
Windows 11 Pro 26200
MSI Center: Extreme Performance mode, Discrete Graphics Mode
Problem:
When the RTX 5080 eGPU is connected, NPCF (nvpcf.sys) assigns Dynamic Boost to the external desktop GPU instead of the internal laptop GPU. This locks the RTX 5090 Laptop at its base TDP of 95W instead of 175W.

Testing performed:

Condition 5090 enforced power limit
eGPU disabled in Device Manager 175W (correct)
eGPU enabled 95W (incorrect)
eGPU enabled + NPCF disabled 95W (no Dynamic Boost at all)
eGPU enabled + NPCF cycled (disable/re-enable) 95W
eGPU enabled + MSI Center profile toggled (Extreme/Balanced/ECO via WMI) 95W
nvidia-smi -pl 175 (admin) "Changing power management limit is not supported"
NVML nvmlDeviceSetPowerManagementLimit "Not Supported"
Conclusion:
NPCF is the sole source of Dynamic Boost for the laptop GPU. When two NVIDIA GPUs are present, it targets the wrong one. The desktop eGPU has its own 360W power supply and does not need Dynamic Boost.

Expected behavior:
nvpcf.sys should identify the internal laptop GPU by:

PCI bus topology (01:00.0 = internal, 54:00.0 = Thunderbolt)
Subsystem vendor ID (1462 = MSI OEM = internal GPU)
Or ACPI _DSM presence
and assign Dynamic Boost exclusively to the internal GPU.

Workaround:
The only working workaround is disabling the eGPU in Device Manager, which defeats the purpose of having one.

This issue also affects RTX 40-series laptops — see egpu.io thread and GeForce Forums thread.